Traditional NLP

[abstractqqq]

0. Stemming. Snowball stemming. Only English.
1. Count Vectorizer, mostly done
2. TFIDF Vectorizer, mostly done
3. Text related metrics, cosine_similarity, Levenshtein distance are done. Some others? Like Jaccard similarity for sets of text.
4. Naive Bayes Classification.
5. Latent Semantic Analysis. Need more ndarray + nalgebra integration with Rust.
6. Other generic text cleaning/processing.